/[pcre]/code/trunk/ChangeLog
ViewVC logotype

Diff of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log | View Patch Patch

revision 31 by nigel, Sat Feb 24 21:38:57 2007 UTC revision 175 by ph10, Mon Jun 11 13:38:38 2007 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
3    
4    Version 7.2 05-June-07
5    ---------------------
6    
7     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
8        which is apparently normally available under Windows.
9    
10     2. Re-jig the pcregrep tests with different newline settings in an attempt
11        to make them independent of the local environment's newline setting.
12    
13     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
14    
15     4. Some of the "internals" tests were previously cut out when the link size
16        was not 2, because the output contained actual offsets. The recent new
17        "Z" feature of pcretest means that these can be cut out, making the tests
18        usable with all link sizes.
19    
20     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
21        stack recursion. This gives a massive performance boost under BSD, but just
22        a small improvement under Linux. However, it saves one field in the frame
23        in all cases.
24    
25     6. Added more features from the forthcoming Perl 5.10:
26    
27        (a) (?-n) (where n is a string of digits) is a relative subroutine or
28            recursion call. It refers to the nth most recently opened parentheses.
29    
30        (b) (?+n) is also a relative subroutine call; it refers to the nth next
31            to be opened parentheses.
32    
33        (c) Conditions that refer to capturing parentheses can be specified
34            relatively, for example, (?(-2)... or (?(+3)...
35    
36        (d) \K resets the start of the current match so that everything before
37            is not part of it.
38    
39        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
40    
41        (f) \g{name} is another synonym - part of Perl 5.10's unification of
42            reference syntax.
43    
44        (g) (?| introduces a group in which the numbering of parentheses in each
45            alternative starts with the same number.
46    
47     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
48        PCRE_INFO_JCHANGED.
49    
50     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
51        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
52        for detecting groups that can match an empty string.
53    
54     9. A pattern with a very large number of alternatives (more than several
55        hundred) was running out of internal workspace during the pre-compile
56        phase, where pcre_compile() figures out how much memory will be needed. A
57        bit of new cunning has reduced the workspace needed for groups with
58        alternatives. The 1000-alternative test pattern now uses 12 bytes of
59        workspace instead of running out of the 4096 that are available.
60    
61    
62    Version 7.1 24-Apr-07
63    ---------------------
64    
65     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
66        that is more "standard", making use of automake and other Autotools. There
67        is some re-arrangement of the files and adjustment of comments consequent
68        on this.
69    
70     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
71        for recursive directory scanning broke on some systems because the files
72        are not scanned in any specific order and on different systems the order
73        was different. A call to "sort" has been inserted into RunGrepTest for the
74        approprate test as a short-term fix. In the longer term there may be an
75        alternative.
76    
77     3. I had an email from Eric Raymond about problems translating some of PCRE's
78        man pages to HTML (despite the fact that I distribute HTML pages, some
79        people do their own conversions for various reasons). The problems
80        concerned the use of low-level troff macros .br and .in. I have therefore
81        removed all such uses from the man pages (some were redundant, some could
82        be replaced by .nf/.fi pairs). The 132html script that I use to generate
83        HTML has been updated to handle .nf/.fi and to complain if it encounters
84        .br or .in.
85    
86     4. Updated comments in configure.ac that get placed in config.h.in and also
87        arranged for config.h to be included in the distribution, with the name
88        config.h.generic, for the benefit of those who have to compile without
89        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
90    
91     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
92        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
93        makevp.bat for the current PCRE, using the additional files
94        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
95    
96     6. A Windows user reported a minor discrepancy with test 2, which turned out
97        to be caused by a trailing space on an input line that had got lost in his
98        copy. The trailing space was an accident, so I've just removed it.
99    
100     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
101        that is needed.
102    
103     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
104        as "const" (a) because they are and (b) because it helps the PHP
105        maintainers who have recently made a script to detect big data structures
106        in the php code that should be moved to the .rodata section. I remembered
107        to update Builducptable as well, so it won't revert if ucptable.h is ever
108        re-created.
109    
110     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
111        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
112        order to be able to cut out the UTF-8 tables in the latter when UTF-8
113        support is not required. This saves 1.5-2K of code, which is important in
114        some applications.
115    
116        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
117        so as not to refer to the tables, even though these functions will never be
118        called when UTF-8 support is disabled. Otherwise there are problems with a
119        shared library.
120    
121    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
122    
123        (a) It was defining its arguments as char * instead of void *.
124    
125        (b) It was assuming that all moves were upwards in memory; this was true
126            a long time ago when I wrote it, but is no longer the case.
127    
128        The emulated memove() is provided for those environments that have neither
129        memmove() nor bcopy(). I didn't think anyone used it these days, but that
130        is clearly not the case, as these two bugs were recently reported.
131    
132    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
133        and Detrail to create the HTML documentation, the .txt form of the man
134        pages, and it removes trailing spaces from listed files. It also creates
135        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
136        case, it wraps all the #defines with #ifndefs. This script should be run
137        before "make dist".
138    
139    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
140        with Unicode property support.
141    
142        (a) For a maximizing quantifier, if the two different cases of the
143            character were of different lengths in their UTF-8 codings (there are
144            some cases like this - I found 11), and the matching function had to
145            back up over a mixture of the two cases, it incorrectly assumed they
146            were both the same length.
147    
148        (b) When PCRE was configured to use the heap rather than the stack for
149            recursion during matching, it was not correctly preserving the data for
150            the other case of a UTF-8 character when checking ahead for a match
151            while processing a minimizing repeat. If the check also involved
152            matching a wide character, but failed, corruption could cause an
153            erroneous result when trying to check for a repeat of the original
154            character.
155    
156    13. Some tidying changes to the testing mechanism:
157    
158        (a) The RunTest script now detects the internal link size and whether there
159            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
160            values substituted by "configure". (The RunGrepTest script already did
161            this for UTF-8.) The configure.ac script no longer substitutes the
162            relevant variables.
163    
164        (b) The debugging options /B and /D in pcretest show the compiled bytecode
165            with length and offset values. This means that the output is different
166            for different internal link sizes. Test 2 is skipped for link sizes
167            other than 2 because of this, bypassing the problem. Unfortunately,
168            there was also a test in test 3 (the locale tests) that used /B and
169            failed for link sizes other than 2. Rather than cut the whole test out,
170            I have added a new /Z option to pcretest that replaces the length and
171            offset values with spaces. This is now used to make test 3 independent
172            of link size. (Test 2 will be tidied up later.)
173    
174    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
175        segmentation fault instead of returning the appropriate error message.
176    
177    15. In multiline mode when the newline sequence was set to "any", the pattern
178        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
179        This doesn't seem right; it now treats the CRLF combination as the line
180        ending, and so does not match in that case. It's only a pattern such as ^$
181        that would hit this one: something like ^ABC$ would have failed after \r
182        and then tried again after \r\n.
183    
184    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
185        in an attempt to make files that differ only in their line terminators
186        compare equal. This works on Linux.
187    
188    17. Under certain error circumstances pcregrep might try to free random memory
189        as it exited. This is now fixed, thanks to valgrind.
190    
191    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
192        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
193        was because its rules for how to advance for /g after matching an empty
194        string at the end of a line did not allow for this case. They now check for
195        it specially.
196    
197    20. pcretest is supposed to handle patterns and data of any length, by
198        extending its buffers when necessary. It was getting this wrong when the
199        buffer for a data line had to be extended.
200    
201    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
202        CRLF as a newline sequence.
203    
204    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
205        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
206        I have nevertheless tidied it up.
207    
208    23. Added some casts to kill warnings from HP-UX ia64 compiler.
209    
210    24. Added a man page for pcre-config.
211    
212    
213    Version 7.0 19-Dec-06
214    ---------------------
215    
216     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
217        moving to gcc 4.1.1.
218    
219     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
220        sys/time.h, which is documented as needed for this function. It doesn't
221        seem to matter on Linux, but it showed up on some releases of OS X.
222    
223     3. It seems that there are systems where bytes whose values are greater than
224        127 match isprint() in the "C" locale. The "C" locale should be the
225        default when a C program starts up. In most systems, only ASCII printing
226        characters match isprint(). This difference caused the output from pcretest
227        to vary, making some of the tests fail. I have changed pcretest so that:
228    
229        (a) When it is outputting text in the compiled version of a pattern, bytes
230            other than 32-126 are always shown as hex escapes.
231    
232        (b) When it is outputting text that is a matched part of a subject string,
233            it does the same, unless a different locale has been set for the match
234            (using the /L modifier). In this case, it uses isprint() to decide.
235    
236     4. Fixed a major bug that caused incorrect computation of the amount of memory
237        required for a compiled pattern when options that changed within the
238        pattern affected the logic of the preliminary scan that determines the
239        length. The relevant options are -x, and -i in UTF-8 mode. The result was
240        that the computed length was too small. The symptoms of this bug were
241        either the PCRE error "internal error: code overflow" from pcre_compile(),
242        or a glibc crash with a message such as "pcretest: free(): invalid next
243        size (fast)". Examples of patterns that provoked this bug (shown in
244        pcretest format) are:
245    
246          /(?-x: )/x
247          /(?x)(?-x: \s*#\s*)/
248          /((?i)[\x{c0}])/8
249          /(?i:[\x{c0}])/8
250    
251        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
252        is now done differently.
253    
254     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
255        wrapper classes; (b) implement a new function in the C++ scanner that is
256        more efficient than the old way of doing things because it avoids levels of
257        recursion in the regex matching; (c) add a paragraph to the documentation
258        for the FullMatch() function.
259    
260     6. The escape sequence \n was being treated as whatever was defined as
261        "newline". Not only was this contrary to the documentation, which states
262        that \n is character 10 (hex 0A), but it also went horribly wrong when
263        "newline" was defined as CRLF. This has been fixed.
264    
265     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
266        was being set to -1 for the "end of line" case (supposedly a value that no
267        character can have). Though this value is never used (the check for end of
268        line is "zero bytes in current character"), it caused compiler complaints.
269        I've changed it to 0xffffffff.
270    
271     8. In pcre_version.c, the version string was being built by a sequence of
272        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
273        string (as it is for production releases) called a macro with an empty
274        argument. The C standard says the result of this is undefined. The gcc
275        compiler treats it as an empty string (which was what was wanted) but it is
276        reported that Visual C gives an error. The source has been hacked around to
277        avoid this problem.
278    
279     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
280        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
281        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
282        of them did).
283    
284    10. Originally, pcretest opened its input and output without "b"; then I was
285        told that "b" was needed in some environments, so it was added for release
286        5.0 to both the input and output. (It makes no difference on Unix-like
287        systems.) Later I was told that it is wrong for the input on Windows. I've
288        now abstracted the modes into two macros, to make it easier to fiddle with
289        them, and removed "b" from the input mode under Windows.
290    
291    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
292    
293    12. Added -help and --help to pcretest as an official way of being reminded
294        of the options.
295    
296    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
297        and pcrecpp.cc because they annoy compilers at high warning levels.
298    
299    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
300    
301    15. Fixed an occurrence of == in configure.ac that should have been = (shell
302        scripts are not C programs :-) and which was not noticed because it works
303        on Linux.
304    
305    16. pcretest is supposed to handle any length of pattern and data line (as one
306        line or as a continued sequence of lines) by extending its input buffer if
307        necessary. This feature was broken for very long pattern lines, leading to
308        a string of junk being passed to pcre_compile() if the pattern was longer
309        than about 50K.
310    
311    17. I have done a major re-factoring of the way pcre_compile() computes the
312        amount of memory needed for a compiled pattern. Previously, there was code
313        that made a preliminary scan of the pattern in order to do this. That was
314        OK when PCRE was new, but as the facilities have expanded, it has become
315        harder and harder to keep it in step with the real compile phase, and there
316        have been a number of bugs (see for example, 4 above). I have now found a
317        cunning way of running the real compile function in a "fake" mode that
318        enables it to compute how much memory it would need, while actually only
319        ever using a few hundred bytes of working memory and without too many
320        tests of the mode. This should make future maintenance and development
321        easier. A side effect of this work is that the limit of 200 on the nesting
322        depth of parentheses has been removed (though this was never a serious
323        limitation, I suspect). However, there is a downside: pcre_compile() now
324        runs more slowly than before (30% or more, depending on the pattern). I
325        hope this isn't a big issue. There is no effect on runtime performance.
326    
327    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
328        newline (only possible for the last line of a file) and it was a
329        pattern that set a locale (followed by /Lsomething), pcretest crashed.
330    
331    19. Added additional timing features to pcretest. (1) The -tm option now times
332        matching only, not compiling. (2) Both -t and -tm can be followed, as a
333        separate command line item, by a number that specifies the number of
334        repeats to use when timing. The default is 50000; this gives better
335        precision, but takes uncomfortably long for very large patterns.
336    
337    20. Extended pcre_study() to be more clever in cases where a branch of a
338        subpattern has no definite first character. For example, (a*|b*)[cd] would
339        previously give no result from pcre_study(). Now it recognizes that the
340        first character must be a, b, c, or d.
341    
342    21. There was an incorrect error "recursive call could loop indefinitely" if
343        a subpattern (or the entire pattern) that was being tested for matching an
344        empty string contained only one non-empty item after a nested subpattern.
345        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
346        incorrectly, because the \d was being skipped in the check.
347    
348    22. The pcretest program now has a new pattern option /B and a command line
349        option -b, which is equivalent to adding /B to every pattern. This causes
350        it to show the compiled bytecode, without the additional information that
351        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
352        is the same as /B/I).
353    
354    23. A new optimization is now able automatically to treat some sequences such
355        as a*b as a*+b. More specifically, if something simple (such as a character
356        or a simple class like \d) has an unlimited quantifier, and is followed by
357        something that cannot possibly match the quantified thing, the quantifier
358        is automatically "possessified".
359    
360    24. A recursive reference to a subpattern whose number was greater than 39
361        went wrong under certain circumstances in UTF-8 mode. This bug could also
362        have affected the operation of pcre_study().
363    
364    25. Realized that a little bit of performance could be had by replacing
365        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
366    
367    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
368    
369    27. Possessive quantifiers such as a++ were previously implemented by turning
370        them into atomic groups such as ($>a+). Now they have their own opcodes,
371        which improves performance. This includes the automatically created ones
372        from 23 above.
373    
374    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
375        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
376        the first matched character to be a colon. This applied both to named and
377        numbered groups.
378    
379    29. The ucpinternal.h header file was missing its idempotency #ifdef.
380    
381    30. I was sent a "project" file called libpcre.a.dev which I understand makes
382        building PCRE on Windows easier, so I have included it in the distribution.
383    
384    31. There is now a check in pcretest against a ridiculously large number being
385        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
386        loop, the loop is abandoned.
387    
388    32. Forward references to subpatterns in conditions such as (?(2)...) where
389        subpattern 2 is defined later cause pcre_compile() to search forwards in
390        the pattern for the relevant set of parentheses. This search went wrong
391        when there were unescaped parentheses in a character class, parentheses
392        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
393    
394    33. "Subroutine" calls and backreferences were previously restricted to
395        referencing subpatterns earlier in the regex. This restriction has now
396        been removed.
397    
398    34. Added a number of extra features that are going to be in Perl 5.10. On the
399        whole, these are just syntactic alternatives for features that PCRE had
400        previously implemented using the Python syntax or my own invention. The
401        other formats are all retained for compatibility.
402    
403        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
404            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
405            also .NET compatible.
406    
407        (b) A recursion or subroutine call to a named group can now be defined as
408            (?&name) as well as (?P>name).
409    
410        (c) A backreference to a named group can now be defined as \k<name> or
411            \k'name' as well as (?P=name). The new forms, as well as being in Perl
412            5.10, are also .NET compatible.
413    
414        (d) A conditional reference to a named group can now use the syntax
415            (?(<name>) or (?('name') as well as (?(name).
416    
417        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
418            groups (named and numbered) that are never evaluated inline, but can be
419            called as "subroutines" from elsewhere. In effect, the DEFINE condition
420            is always false. There may be only one alternative in such a group.
421    
422        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
423            as the simple (?(R). The condition is true only if the most recent
424            recursion is that of the given number or name. It does not search out
425            through the entire recursion stack.
426    
427        (g) The escape \gN or \g{N} has been added, where N is a positive or
428            negative number, specifying an absolute or relative reference.
429    
430    35. Tidied to get rid of some further signed/unsigned compiler warnings and
431        some "unreachable code" warnings.
432    
433    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
434        things, this adds five new scripts.
435    
436    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
437        There were also incompatibilities regarding the handling of \Q..\E inside
438        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
439        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
440    
441    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
442        matches an empty string, and forcibly breaks the loop. There were bugs in
443        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
444        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
445        separate and independent bugs (that affected different cases) have been
446        fixed.
447    
448    39. Refactored the code to abolish the use of different opcodes for small
449        capturing bracket numbers. This is a tidy that I avoided doing when I
450        removed the limit on the number of capturing brackets for 3.5 back in 2001.
451        The new approach is not only tidier, it makes it possible to reduce the
452        memory needed to fix the previous bug (38).
453    
454    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
455        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
456        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
457        mode.
458    
459    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
460        report.
461    
462    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
463        copy construction and assignment in the C++ wrapper.
464    
465    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
466        couple of bugs that could have given wrong results in the "--newline=crlf"
467        case.
468    
469    44. Added a number of casts and did some reorganization of signed/unsigned int
470        variables following suggestions from Dair Grant. Also renamed the variable
471        "this" as "item" because it is a C++ keyword.
472    
473    45. Arranged for dftables to add
474    
475          #include "pcre_internal.h"
476    
477        to pcre_chartables.c because without it, gcc 4.x may remove the array
478        definition from the final binary if PCRE is built into a static library and
479        dead code stripping is activated.
480    
481    46. For an unanchored pattern, if a match attempt fails at the start of a
482        newline sequence, and the newline setting is CRLF or ANY, and the next two
483        characters are CRLF, advance by two characters instead of one.
484    
485    
486    Version 6.7 04-Jul-06
487    ---------------------
488    
489     1. In order to handle tests when input lines are enormously long, pcretest has
490        been re-factored so that it automatically extends its buffers when
491        necessary. The code is crude, but this _is_ just a test program. The
492        default size has been increased from 32K to 50K.
493    
494     2. The code in pcre_study() was using the value of the re argument before
495        testing it for NULL. (Of course, in any sensible call of the function, it
496        won't be NULL.)
497    
498     3. The memmove() emulation function in pcre_internal.h, which is used on
499        systems that lack both memmove() and bcopy() - that is, hardly ever -
500        was missing a "static" storage class specifier.
501    
502     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
503        containing an extended class (one that cannot be represented by a bitmap
504        because it contains high-valued characters or Unicode property items, e.g.
505        [\pZ]). Almost always one would set UTF-8 mode when processing such a
506        pattern, but PCRE should not loop if you do not (it no longer does).
507        [Detail: two cases were found: (a) a repeated subpattern containing an
508        extended class; (b) a recursive reference to a subpattern that followed a
509        previous extended class. It wasn't skipping over the extended class
510        correctly when UTF-8 mode was not set.]
511    
512     5. A negated single-character class was not being recognized as fixed-length
513        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
514        compile error "lookbehind assertion is not fixed length".
515    
516     6. The RunPerlTest auxiliary script was showing an unexpected difference
517        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
518        write a Perl script that can interpret lines of an input file either as
519        byte characters or as UTF-8, which is what "perltest" was being required to
520        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
521        can't do is switch easily at run time between having the "use utf8;" pragma
522        or not. In the end, I fudged it by using the RunPerlTest script to insert
523        "use utf8;" explicitly for the UTF-8 tests.
524    
525     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
526        the end of the subject string, contrary to the documentation and to what
527        Perl does. This was true of both matching functions. Now it matches only at
528        the start of the subject and immediately after *internal* newlines.
529    
530     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
531        a pointer to an int instead of a pointer to an unsigned long int. This
532        caused problems on 64-bit systems.
533    
534     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
535        instance of the 'standard' template library not being so standard".
536    
537    10. There was no check on the number of named subpatterns nor the maximum
538        length of a subpattern name. The product of these values is used to compute
539        the size of the memory block for a compiled pattern. By supplying a very
540        long subpattern name and a large number of named subpatterns, the size
541        computation could be caused to overflow. This is now prevented by limiting
542        the length of names to 32 characters, and the number of named subpatterns
543        to 10,000.
544    
545    11. Subpatterns that are repeated with specific counts have to be replicated in
546        the compiled pattern. The size of memory for this was computed from the
547        length of the subpattern and the repeat count. The latter is limited to
548        65535, but there was no limit on the former, meaning that integer overflow
549        could in principle occur. The compiled length of a repeated subpattern is
550        now limited to 30,000 bytes in order to prevent this.
551    
552    12. Added the optional facility to have named substrings with the same name.
553    
554    13. Added the ability to use a named substring as a condition, using the
555        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
556        are numbers (not recommended). Forward references are permitted.
557    
558    14. Added forward references in named backreferences (if you see what I mean).
559    
560    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
561        pattern could run off the end of the subject. For example, the pattern
562        "(?s)(.{1,5})"8 did this with the subject "ab".
563    
564    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
565        PCRE_CASELESS was set when matching characters that were quantified with ?
566        or *.
567    
568    17. A character class other than a single negated character that had a minimum
569        but no maximum quantifier - for example [ab]{6,} - was not handled
570        correctly by pce_dfa_exec(). It would match only one character.
571    
572    18. A valid (though odd) pattern that looked like a POSIX character
573        class but used an invalid character after [ (for example [[,abc,]]) caused
574        pcre_compile() to give the error "Failed: internal error: code overflow" or
575        in some cases to crash with a glibc free() error. This could even happen if
576        the pattern terminated after [[ but there just happened to be a sequence of
577        letters, a binary zero, and a closing ] in the memory that followed.
578    
579    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
580        over the years. Originally (before any Unicode support), just the bottom 8
581        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
582        output from "man perlunicode" includes this:
583    
584          The regular expression compiler produces polymorphic opcodes.  That
585          is, the pattern adapts to the data and automatically switches to
586          the Unicode character scheme when presented with Unicode data--or
587          instead uses a traditional byte scheme when presented with byte
588          data.
589    
590        Sadly, a wide octal escape does not cause a switch, and in a string with
591        no other multibyte characters, these octal escapes are treated as before.
592        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
593        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
594        Unicode string.
595    
596        I have not perpetrated such confusion in PCRE. Up till now, it took just
597        the bottom 8 bits, as in old Perl. I have now made octal escapes with
598        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
599        translate to the appropriate multibyte character.
600    
601    29. Applied some refactoring to reduce the number of warnings from Microsoft
602        and Borland compilers. This has included removing the fudge introduced
603        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
604        a warning about an unused variable.
605    
606    21. PCRE has not included VT (character 0x0b) in the set of whitespace
607        characters since release 4.0, because Perl (from release 5.004) does not.
608        [Or at least, is documented not to: some releases seem to be in conflict
609        with the documentation.] However, when a pattern was studied with
610        pcre_study() and all its branches started with \s, PCRE still included VT
611        as a possible starting character. Of course, this did no harm; it just
612        caused an unnecessary match attempt.
613    
614    22. Removed a now-redundant internal flag bit that recorded the fact that case
615        dependency changed within the pattern. This was once needed for "required
616        byte" processing, but is no longer used. This recovers a now-scarce options
617        bit. Also moved the least significant internal flag bit to the most-
618        significant bit of the word, which was not previously used (hangover from
619        the days when it was an int rather than a uint) to free up another bit for
620        the future.
621    
622    23. Added support for CRLF line endings as well as CR and LF. As well as the
623        default being selectable at build time, it can now be changed at runtime
624        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
625        specify that it is scanning data with non-default line endings.
626    
627    24. Changed the definition of CXXLINK to make it agree with the definition of
628        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
629    
630    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
631        recursions. This makes a big different to stack usage for some patterns.
632    
633    26. If a subpattern containing a named recursion or subroutine reference such
634        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
635        the space required for the compiled pattern went wrong and gave too small a
636        value. Depending on the environment, this could lead to "Failed: internal
637        error: code overflow at offset 49" or "glibc detected double free or
638        corruption" errors.
639    
640    27. Applied patches from Google (a) to support the new newline modes and (b) to
641        advance over multibyte UTF-8 characters in GlobalReplace.
642    
643    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
644        difference for some implementation of PCRE in some Windows version.
645    
646    29. Added some extra testing facilities to pcretest:
647    
648        \q<number>   in a data line sets the "match limit" value
649        \Q<number>   in a data line sets the "match recursion limt" value
650        -S <number>  sets the stack size, where <number> is in megabytes
651    
652        The -S option isn't available for Windows.
653    
654    
655    Version 6.6 06-Feb-06
656    ---------------------
657    
658     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
659        in pcreposix.h. I have copied the definition from pcre.h.
660    
661     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
662        because pcre.h is no longer a built file.
663    
664     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
665        not normally included in the compiled code.
666    
667    
668    Version 6.5 01-Feb-06
669    ---------------------
670    
671     1. When using the partial match feature with pcre_dfa_exec(), it was not
672        anchoring the second and subsequent partial matches at the new starting
673        point. This could lead to incorrect results. For example, with the pattern
674        /1234/, partially matching against "123" and then "a4" gave a match.
675    
676     2. Changes to pcregrep:
677    
678        (a) All non-match returns from pcre_exec() were being treated as failures
679            to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
680            error message is output. Some extra information is given for the
681            PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
682            probably the only errors that are likely to be caused by users (by
683            specifying a regex that has nested indefinite repeats, for instance).
684            If there are more than 20 of these errors, pcregrep is abandoned.
685    
686        (b) A binary zero was treated as data while matching, but terminated the
687            output line if it was written out. This has been fixed: binary zeroes
688            are now no different to any other data bytes.
689    
690        (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
691            used to set a locale for matching. The --locale=xxxx long option has
692            been added (no short equivalent) to specify a locale explicitly on the
693            pcregrep command, overriding the environment variables.
694    
695        (d) When -B was used with -n, some line numbers in the output were one less
696            than they should have been.
697    
698        (e) Added the -o (--only-matching) option.
699    
700        (f) If -A or -C was used with -c (count only), some lines of context were
701            accidentally printed for the final match.
702    
703        (g) Added the -H (--with-filename) option.
704    
705        (h) The combination of options -rh failed to suppress file names for files
706            that were found from directory arguments.
707    
708        (i) Added the -D (--devices) and -d (--directories) options.
709    
710        (j) Added the -F (--fixed-strings) option.
711    
712        (k) Allow "-" to be used as a file name for -f as well as for a data file.
713    
714        (l) Added the --colo(u)r option.
715    
716        (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
717            is not present by default.
718    
719     3. A nasty bug was discovered in the handling of recursive patterns, that is,
720        items such as (?R) or (?1), when the recursion could match a number of
721        alternatives. If it matched one of the alternatives, but subsequently,
722        outside the recursion, there was a failure, the code tried to back up into
723        the recursion. However, because of the way PCRE is implemented, this is not
724        possible, and the result was an incorrect result from the match.
725    
726        In order to prevent this happening, the specification of recursion has
727        been changed so that all such subpatterns are automatically treated as
728        atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
729    
730     4. I had overlooked the fact that, in some locales, there are characters for
731        which isalpha() is true but neither isupper() nor islower() are true. In
732        the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
733        and ordfeminine) are like this. This affected the treatment of \w and \W
734        when they appeared in character classes, but not when they appeared outside
735        a character class. The bit map for "word" characters is now created
736        separately from the results of isalnum() instead of just taking it from the
737        upper, lower, and digit maps. (Plus the underscore character, of course.)
738    
739     5. The above bug also affected the handling of POSIX character classes such as
740        [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
741        permanent tables. Instead, the bit maps for such a class were previously
742        created as the appropriate unions of the upper, lower, and digit bitmaps.
743        Now they are created by subtraction from the [[:word:]] class, which has
744        its own bitmap.
745    
746     6. The [[:blank:]] character class matches horizontal, but not vertical space.
747        It is created by subtracting the vertical space characters (\x09, \x0a,
748        \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
749        subtraction was done in the overall bitmap for a character class, meaning
750        that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
751        be recognized. This bug has been fixed.
752    
753     7. Patches from the folks at Google:
754    
755          (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
756          real life, but is still worth protecting against".
757    
758          (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
759          regular expressions".
760    
761          (c) pcre_scanner.cc: avoid use of std::count() because not all systems
762          have it.
763    
764          (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
765          "configure" and the latter not, in order to fix a problem somebody had
766          with compiling the Arg class on HP-UX.
767    
768          (e) Improve the error-handling of the C++ wrapper a little bit.
769    
770          (f) New tests for checking recursion limiting.
771    
772     8. The pcre_memmove() function, which is used only if the environment does not
773        have a standard memmove() function (and is therefore rarely compiled),
774        contained two bugs: (a) use of int instead of size_t, and (b) it was not
775        returning a result (though PCRE never actually uses the result).
776    
777     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
778        large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
779        returned instead of calling malloc() with an overflowing number that would
780        most likely cause subsequent chaos.
781    
782    10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
783    
784    11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
785        with this option is matched, the nmatch and pmatch options of regexec() are
786        ignored.
787    
788    12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
789        provided in case anyone wants to the the POSIX interface with UTF-8
790        strings.
791    
792    13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
793        C++ linking (needed for some HP-UX environments).
794    
795    14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
796        (unused parameter) and in the pcre_printint() function (omitted "default"
797        switch label when the default is to do nothing).
798    
799    15. Added some code to make it possible, when PCRE is compiled as a C++
800        library, to replace subject pointers for pcre_exec() with a smart pointer
801        class, thus making it possible to process discontinuous strings.
802    
803    16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
804        much the same function. They were added by different people who were trying
805        to make PCRE easy to compile on non-Unix systems. It has been suggested
806        that PCRE_EXPORT be abolished now that there is more automatic apparatus
807        for compiling on Windows systems. I have therefore replaced it with
808        PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
809        defaults to "extern" for C or "extern C" for C++, which works fine on
810        Unix-like systems. It is now possible to override the value of PCRE_DATA_
811        SCOPE with something explicit in config.h. In addition:
812    
813        (a) pcreposix.h still had just "extern" instead of either of these macros;
814            I have replaced it with PCRE_DATA_SCOPE.
815    
816        (b) Functions such as _pcre_xclass(), which are internal to the library,
817            but external in the C sense, all had PCRE_EXPORT in their definitions.
818            This is apparently wrong for the Windows case, so I have removed it.
819            (It makes no difference on Unix-like systems.)
820    
821    17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
822        of recursive calls to match(). This is different to MATCH_LIMIT because
823        that limits the total number of calls to match(), not all of which increase
824        the depth of recursion. Limiting the recursion depth limits the amount of
825        stack (or heap if NO_RECURSE is set) that is used. The default can be set
826        when PCRE is compiled, and changed at run time. A patch from Google adds
827        this functionality to the C++ interface.
828    
829    18. Changes to the handling of Unicode character properties:
830    
831        (a) Updated the table to Unicode 4.1.0.
832    
833        (b) Recognize characters that are not in the table as "Cn" (undefined).
834    
835        (c) I revised the way the table is implemented to a much improved format
836            which includes recognition of ranges. It now supports the ranges that
837            are defined in UnicodeData.txt, and it also amalgamates other
838            characters into ranges. This has reduced the number of entries in the
839            table from around 16,000 to around 3,000, thus reducing its size
840            considerably. I realized I did not need to use a tree structure after
841            all - a binary chop search is just as efficient. Having reduced the
842            number of entries, I extended their size from 6 bytes to 8 bytes to
843            allow for more data.
844    
845        (d) Added support for Unicode script names via properties such as \p{Han}.
846    
847    19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
848        matching that character.
849    
850    20. When matching a repeated Unicode property with a minimum greater than zero,
851        (for example \pL{2,}), PCRE could look past the end of the subject if it
852        reached it while seeking the minimum number of characters. This could
853        happen only if some of the characters were more than one byte long, because
854        there is a check for at least the minimum number of bytes.
855    
856    21. Refactored the implementation of \p and \P so as to be more general, to
857        allow for more different types of property in future. This has changed the
858        compiled form incompatibly. Anybody with saved compiled patterns that use
859        \p or \P will have to recompile them.
860    
861    22. Added "Any" and "L&" to the supported property types.
862    
863    23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
864        but give a compile time error if the value is greater than 0xff.
865    
866    24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
867        accidentally not being installed or uninstalled.
868    
869    25. The pcre.h file was built from pcre.h.in, but the only changes that were
870        made were to insert the current release number. This seemed silly, because
871        it made things harder for people building PCRE on systems that don't run
872        "configure". I have turned pcre.h into a distributed file, no longer built
873        by "configure", with the version identification directly included. There is
874        no longer a pcre.h.in file.
875    
876        However, this change necessitated a change to the pcre-config script as
877        well. It is built from pcre-config.in, and one of the substitutions was the
878        release number. I have updated configure.ac so that ./configure now finds
879        the release number by grepping pcre.h.
880    
881    26. Added the ability to run the tests under valgrind.
882    
883    
884    Version 6.4 05-Sep-05
885    ---------------------
886    
887     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
888        "--" to be printed when multiple files were scanned, even when none of the
889        -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
890        consider it to be a bug, and have restored the previous behaviour.
891    
892     2. A couple of code tidies to get rid of compiler warnings.
893    
894     3. The pcretest program used to cheat by referring to symbols in the library
895        whose names begin with _pcre_. These are internal symbols that are not
896        really supposed to be visible externally, and in some environments it is
897        possible to suppress them. The cheating is now confined to including
898        certain files from the library's source, which is a bit cleaner.
899    
900     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
901        file's purpose clearer.
902    
903     5. Reorganized pcre_ucp_findchar().
904    
905    
906    Version 6.3 15-Aug-05
907    ---------------------
908    
909     1. The file libpcre.pc.in did not have general read permission in the tarball.
910    
911     2. There were some problems when building without C++ support:
912    
913        (a) If C++ support was not built, "make install" and "make test" still
914            tried to test it.
915    
916        (b) There were problems when the value of CXX was explicitly set. Some
917            changes have been made to try to fix these, and ...
918    
919        (c) --disable-cpp can now be used to explicitly disable C++ support.
920    
921        (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
922            backslash in a target when C++ was disabled. This confuses some
923            versions of "make", apparently. Using an intermediate variable solves
924            this. (Same for CPP_LOBJ.)
925    
926     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
927        (non-Windows) now includes $(CFLAGS) because these flags are sometimes
928        necessary on certain architectures.
929    
930     4. Added a setting of -export-symbols-regex to the link command to remove
931        those symbols that are exported in the C sense, but actually are local
932        within the library, and not documented. Their names all begin with
933        "_pcre_". This is not a perfect job, because (a) we have to except some
934        symbols that pcretest ("illegally") uses, and (b) the facility isn't always
935        available (and never for static libraries). I have made a note to try to
936        find a way round (a) in the future.
937    
938    
939    Version 6.2 01-Aug-05
940    ---------------------
941    
942     1. There was no test for integer overflow of quantifier values. A construction
943        such as {1111111111111111} would give undefined results. What is worse, if
944        a minimum quantifier for a parenthesized subpattern overflowed and became
945        negative, the calculation of the memory size went wrong. This could have
946        led to memory overwriting.
947    
948     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
949    
950     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
951        operating environments where this matters.
952    
953     4. Applied Giuseppe Maxia's patch to add additional features for controlling
954        PCRE options from within the C++ wrapper.
955    
956     5. Named capturing subpatterns were not being correctly counted when a pattern
957        was compiled. This caused two problems: (a) If there were more than 100
958        such subpatterns, the calculation of the memory needed for the whole
959        compiled pattern went wrong, leading to an overflow error. (b) Numerical
960        back references of the form \12, where the number was greater than 9, were
961        not recognized as back references, even though there were sufficient
962        previous subpatterns.
963    
964     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
965        versions of gcc, e.g. 2.95.4.
966    
967    
968    Version 6.1 21-Jun-05
969    ---------------------
970    
971     1. There was one reference to the variable "posix" in pcretest.c that was not
972        surrounded by "#if !defined NOPOSIX".
973    
974     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
975        the cross-check on the old pcre_info() function, for the benefit of the
976        cut-down version of PCRE that is currently imported into Exim.
977    
978     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
979        allocation error. I've done the easy fix, which wastes 2 bytes for sensible
980        patterns that start (?i) but I don't think that matters. The use of (?i) is
981        just an example; this all applies to the other options as well.
982    
983     4. Since libtool seems to echo the compile commands it is issuing, the output
984        from "make" can be reduced a bit by putting "@" in front of each libtool
985        compile command.
986    
987     5. Patch from the folks at Google for configure.in to be a bit more thorough
988        in checking for a suitable C++ installation before trying to compile the
989        C++ stuff. This should fix a reported problem when a compiler was present,
990        but no suitable headers.
991    
992     6. The man pages all had just "PCRE" as their title. I have changed them to
993        be the relevant file name. I have also arranged that these names are
994        retained in the file doc/pcre.txt, which is a concatenation in text format
995        of all the man pages except the little individual ones for each function.
996    
997     7. The NON-UNIX-USE file had not been updated for the different set of source
998        files that come with release 6. I also added a few comments about the C++
999        wrapper.
1000    
1001    
1002    Version 6.0 07-Jun-05
1003    ---------------------
1004    
1005     1. Some minor internal re-organization to help with my DFA experiments.
1006    
1007     2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
1008        didn't matter for the library itself when fully configured, but did matter
1009        when compiling without UCP support, or within Exim, where the ucp files are
1010        not imported.
1011    
1012     3. Refactoring of the library code to split up the various functions into
1013        different source modules. The addition of the new DFA matching code (see
1014        below) to a single monolithic source would have made it really too
1015        unwieldy, quite apart from causing all the code to be include in a
1016        statically linked application, when only some functions are used. This is
1017        relevant even without the DFA addition now that patterns can be compiled in
1018        one application and matched in another.
1019    
1020        The downside of splitting up is that there have to be some external
1021        functions and data tables that are used internally in different modules of
1022        the library but which are not part of the API. These have all had their
1023        names changed to start with "_pcre_" so that they are unlikely to clash
1024        with other external names.
1025    
1026     4. Added an alternate matching function, pcre_dfa_exec(), which matches using
1027        a different (DFA) algorithm. Although it is slower than the original
1028        function, it does have some advantages for certain types of matching
1029        problem.
1030    
1031     5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
1032        including restarting after a partial match.
1033    
1034     6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
1035        defined when compiling for Windows was sent to me. I have put it into the
1036        code, though I have no means of testing or verifying it.
1037    
1038     7. Added the pcre_refcount() auxiliary function.
1039    
1040     8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
1041        match before or at the first newline in the subject string. In pcretest,
1042        the /f option on a pattern can be used to set this.
1043    
1044     9. A repeated \w when used in UTF-8 mode with characters greater than 256
1045        would behave wrongly. This has been present in PCRE since release 4.0.
1046    
1047    10. A number of changes to the pcregrep command:
1048    
1049        (a) Refactored how -x works; insert ^(...)$ instead of setting
1050            PCRE_ANCHORED and checking the length, in preparation for adding
1051            something similar for -w.
1052    
1053        (b) Added the -w (match as a word) option.
1054    
1055        (c) Refactored the way lines are read and buffered so as to have more
1056            than one at a time available.
1057    
1058        (d) Implemented a pcregrep test script.
1059    
1060        (e) Added the -M (multiline match) option. This allows patterns to match
1061            over several lines of the subject. The buffering ensures that at least
1062            8K, or the rest of the document (whichever is the shorter) is available
1063            for matching (and similarly the previous 8K for lookbehind assertions).
1064    
1065        (f) Changed the --help output so that it now says
1066    
1067              -w, --word-regex(p)
1068    
1069            instead of two lines, one with "regex" and the other with "regexp"
1070            because that confused at least one person since the short forms are the
1071            same. (This required a bit of code, as the output is generated
1072            automatically from a table. It wasn't just a text change.)
1073    
1074        (g) -- can be used to terminate pcregrep options if the next thing isn't an
1075            option but starts with a hyphen. Could be a pattern or a path name
1076            starting with a hyphen, for instance.
1077    
1078        (h) "-" can be given as a file name to represent stdin.
1079    
1080        (i) When file names are being printed, "(standard input)" is used for
1081            the standard input, for compatibility with GNU grep. Previously
1082            "<stdin>" was used.
1083    
1084        (j) The option --label=xxx can be used to supply a name to be used for
1085            stdin when file names are being printed. There is no short form.
1086    
1087        (k) Re-factored the options decoding logic because we are going to add
1088            two more options that take data. Such options can now be given in four
1089            different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
1090    
1091        (l) Added the -A, -B, and -C options for requesting that lines of context
1092            around matches be printed.
1093    
1094        (m) Added the -L option to print the names of files that do not contain
1095            any matching lines, that is, the complement of -l.
1096    
1097        (n) The return code is 2 if any file cannot be opened, but pcregrep does
1098            continue to scan other files.
1099    
1100        (o) The -s option was incorrectly implemented. For compatibility with other
1101            greps, it now suppresses the error message for a non-existent or non-
1102            accessible file (but not the return code). There is a new option called
1103            -q that suppresses the output of matching lines, which was what -s was
1104            previously doing.
1105    
1106        (p) Added --include and --exclude options to specify files for inclusion
1107            and exclusion when recursing.
1108    
1109    11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
1110        Hopefully, it now does.
1111    
1112    12. Missing cast in pcre_study().
1113    
1114    13. Added an "uninstall" target to the makefile.
1115    
1116    14. Replaced "extern" in the function prototypes in Makefile.in with
1117        "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
1118        world, but is set differently for Windows.
1119    
1120    15. Added a second compiling function called pcre_compile2(). The only
1121        difference is that it has an extra argument, which is a pointer to an
1122        integer error code. When there is a compile-time failure, this is set
1123        non-zero, in addition to the error test pointer being set to point to an
1124        error message. The new argument may be NULL if no error number is required
1125        (but then you may as well call pcre_compile(), which is now just a
1126        wrapper). This facility is provided because some applications need a
1127        numeric error indication, but it has also enabled me to tidy up the way
1128        compile-time errors are handled in the POSIX wrapper.
1129    
1130    16. Added VPATH=.libs to the makefile; this should help when building with one
1131        prefix path and installing with another. (Or so I'm told by someone who
1132        knows more about this stuff than I do.)
1133    
1134    17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
1135        passes PCRE_DOTALL to the pcre_compile() function, making the "." character
1136        match everything, including newlines. This is not POSIX-compatible, but
1137        somebody wanted the feature. From pcretest it can be activated by using
1138        both the P and the s flags.
1139    
1140    18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
1141    
1142    19. libpcre.pc was being incorrectly installed as executable.
1143    
1144    20. A couple of places in pcretest check for end-of-line by looking for '\n';
1145        it now also looks for '\r' so that it will work unmodified on Windows.
1146    
1147    21. Added Google's contributed C++ wrapper to the distribution.
1148    
1149    22. Added some untidy missing memory free() calls in pcretest, to keep
1150        Electric Fence happy when testing.
1151    
1152    
1153    
1154    Version 5.0 13-Sep-04
1155    ---------------------
1156    
1157     1. Internal change: literal characters are no longer packed up into items
1158        containing multiple characters in a single byte-string. Each character
1159        is now matched using a separate opcode. However, there may be more than one
1160        byte in the character in UTF-8 mode.
1161    
1162     2. The pcre_callout_block structure has two new fields: pattern_position and
1163        next_item_length. These contain the offset in the pattern to the next match
1164        item, and its length, respectively.
1165    
1166     3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
1167        insertion of callouts before each pattern item. Added the /C option to
1168        pcretest to make use of this.
1169    
1170     4. On the advice of a Windows user, the lines
1171    
1172          #if defined(_WIN32) || defined(WIN32)
1173          _setmode( _fileno( stdout ), 0x8000 );
1174          #endif  /* defined(_WIN32) || defined(WIN32) */
1175    
1176        have been added to the source of pcretest. This apparently does useful
1177        magic in relation to line terminators.
1178    
1179     5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
1180        for the benefit of those environments where the "b" makes a difference.
1181    
1182     6. The icc compiler has the same options as gcc, but "configure" doesn't seem
1183        to know about it. I have put a hack into configure.in that adds in code
1184        to set GCC=yes if CC=icc. This seems to end up at a point in the
1185        generated configure script that is early enough to affect the setting of
1186        compiler options, which is what is needed, but I have no means of testing
1187        whether it really works. (The user who reported this had patched the
1188        generated configure script, which of course I cannot do.)
1189    
1190        LATER: After change 22 below (new libtool files), the configure script
1191        seems to know about icc (and also ecc). Therefore, I have commented out
1192        this hack in configure.in.
1193    
1194     7. Added support for pkg-config (2 patches were sent in).
1195    
1196     8. Negated POSIX character classes that used a combination of internal tables
1197        were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
1198        [[:^ascii]]. Typically, they would match almost any characters. The other
1199        POSIX classes were not broken in this way.
1200    
1201     9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
1202        to find the match, as PCRE was deluded into thinking that the match had to
1203        start at the start point or following a newline. The same bug applied to
1204        patterns with negative forward assertions or any backward assertions
1205        preceding ".*" at the start, unless the pattern required a fixed first
1206        character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
1207    
1208    10. In UTF-8 mode, when moving forwards in the subject after a failed match
1209        starting at the last subject character, bytes beyond the end of the subject
1210        string were read.
1211    
1212    11. Renamed the variable "class" as "classbits" to make life easier for C++
1213        users. (Previously there was a macro definition, but it apparently wasn't
1214        enough.)
1215    
1216    12. Added the new field "tables" to the extra data so that tables can be passed
1217        in at exec time, or the internal tables can be re-selected. This allows
1218        a compiled regex to be saved and re-used at a later time by a different
1219        program that might have everything at different addresses.
1220    
1221    13. Modified the pcre-config script so that, when run on Solaris, it shows a
1222        -R library as well as a -L library.
1223    
1224    14. The debugging options of pcretest (-d on the command line or D on a
1225        pattern) showed incorrect output for anything following an extended class
1226        that contained multibyte characters and which was followed by a quantifier.
1227    
1228    15. Added optional support for general category Unicode character properties
1229        via the \p, \P, and \X escapes. Unicode property support implies UTF-8
1230        support. It adds about 90K to the size of the library. The meanings of the
1231        inbuilt class escapes such as \d and \s have NOT been changed.
1232    
1233    16. Updated pcredemo.c to include calls to free() to release the memory for the
1234        compiled pattern.
1235    
1236    17. The generated file chartables.c was being created in the source directory
1237        instead of in the building directory. This caused the build to fail if the
1238        source directory was different from the building directory, and was
1239        read-only.
1240    
1241    18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
1242        file. No doubt somebody will tell me if they don't make sense... Also added
1243        Dan Mooney's comments about building on OpenVMS.
1244    
1245    19. Added support for partial matching via the PCRE_PARTIAL option for
1246        pcre_exec() and the \P data escape in pcretest.
1247    
1248    20. Extended pcretest with 3 new pattern features:
1249    
1250        (i)   A pattern option of the form ">rest-of-line" causes pcretest to
1251              write the compiled pattern to the file whose name is "rest-of-line".
1252              This is a straight binary dump of the data, with the saved pointer to
1253              the character tables forced to be NULL. The study data, if any, is
1254              written too. After writing, pcretest reads a new pattern.
1255    
1256        (ii)  If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
1257              compiled pattern from the given file. There must not be any
1258              occurrences of "<" in the file name (pretty unlikely); if there are,
1259              pcretest will instead treat the initial "<" as a pattern delimiter.
1260              After reading in the pattern, pcretest goes on to read data lines as
1261              usual.
1262    
1263        (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
1264              and 16-bit fields in a compiled pattern, to simulate a pattern that
1265              was compiled on a host of opposite endianness.
1266    
1267    21. The pcre-exec() function can now cope with patterns that were compiled on
1268        hosts of opposite endianness, with this restriction:
1269    
1270          As for any compiled expression that is saved and used later, the tables
1271          pointer field cannot be preserved; the extra_data field in the arguments
1272          to pcre_exec() should be used to pass in a tables address if a value
1273          other than the default internal tables were used at compile time.
1274    
1275    22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
1276        now diagnosed as an error. Previously, most of the time, a negative number
1277        would have been treated as zero, but if in addition "ovector" was passed as
1278        NULL, a crash could occur.
1279    
1280    23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
1281        new versions from the libtool 1.5 distribution (the last one is a copy of
1282        a file called libtool.m4). This seems to have fixed the need to patch
1283        "configure" to support Darwin 1.3 (which I used to do). However, I still
1284        had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
1285        workstation).
1286    
1287    24. Changed the PCRE licence to be the more standard "BSD" licence.
1288    
1289    
1290    Version 4.5 01-Dec-03
1291    ---------------------
1292    
1293     1. There has been some re-arrangement of the code for the match() function so
1294        that it can be compiled in a version that does not call itself recursively.
1295        Instead, it keeps those local variables that need separate instances for
1296        each "recursion" in a frame on the heap, and gets/frees frames whenever it
1297        needs to "recurse". Keeping track of where control must go is done by means
1298        of setjmp/longjmp. The whole thing is implemented by a set of macros that
1299        hide most of the details from the main code, and operates only if
1300        NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
1301        "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
1302        operating.
1303    
1304        To make it easier for callers to provide specially tailored get/free
1305        functions for this usage, two new functions, pcre_stack_malloc, and
1306        pcre_stack_free, are used. They are always called in strict stacking order,
1307        and the size of block requested is always the same.
1308    
1309        The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
1310        PCRE has been compiled to use the stack or the heap for recursion. The
1311        -C option of pcretest uses this to show which version is compiled.
1312    
1313        A new data escape \S, is added to pcretest; it causes the amounts of store
1314        obtained and freed by both kinds of malloc/free at match time to be added
1315        to the output.
1316    
1317     2. Changed the locale test to use "fr_FR" instead of "fr" because that's
1318        what's available on my current Linux desktop machine.
1319    
1320     3. When matching a UTF-8 string, the test for a valid string at the start has
1321        been extended. If start_offset is not zero, PCRE now checks that it points
1322        to a byte that is the start of a UTF-8 character. If not, it returns
1323        PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
1324        this is necessary because there may be backward assertions in the pattern.
1325        When matching the same subject several times, it may save resources to use
1326        PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
1327    
1328     4. The code for checking the validity of UTF-8 strings has been tightened so
1329        that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
1330        containing "overlong sequences".
1331    
1332     5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
1333        I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
1334        should have been "&", but it just so happened that all the cases this let
1335        through by mistake were picked up later in the function.
1336    
1337     6. I had used a variable called "isblank" - this is a C99 function, causing
1338        some compilers to warn. To avoid this, I renamed it (as "blankclass").
1339    
1340     7. Cosmetic: (a) only output another newline at the end of pcretest if it is
1341        prompting; (b) run "./pcretest /dev/null" at the start of the test script
1342        so the version is shown; (c) stop "make test" echoing "./RunTest".
1343    
1344     8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
1345    
1346     9. The prototype for memmove() for systems that don't have it was using
1347        size_t, but the inclusion of the header that defines size_t was later. I've
1348        moved the #includes for the C headers earlier to avoid this.
1349    
1350    10. Added some adjustments to the code to make it easier to compiler on certain
1351        special systems:
1352    
1353          (a) Some "const" qualifiers were missing.
1354          (b) Added the macro EXPORT before all exported functions; by default this
1355              is defined to be empty.
1356          (c) Changed the dftables auxiliary program (that builds chartables.c) so
1357              that it reads its output file name as an argument instead of writing
1358              to the standard output and assuming this can be redirected.
1359    
1360    11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
1361        class containing characters with values greater than 255, PCRE compilation
1362        went into a loop.
1363    
1364    12. A recursive reference to a subpattern that was within another subpattern
1365        that had a minimum quantifier of zero caused PCRE to crash. For example,
1366        (x(y(?2))z)? provoked this bug with a subject that got as far as the
1367        recursion. If the recursively-called subpattern itself had a zero repeat,
1368        that was OK.
1369    
1370    13. In pcretest, the buffer for reading a data line was set at 30K, but the
1371        buffer into which it was copied (for escape processing) was still set at
1372        1024, so long lines caused crashes.
1373    
1374    14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
1375        "internal error: code overflow...". This applied to any character class
1376        that was followed by a possessive quantifier.
1377    
1378    15. Modified the Makefile to add libpcre.la as a prerequisite for
1379        libpcreposix.la because I was told this is needed for a parallel build to
1380        work.
1381    
1382    16. If a pattern that contained .* following optional items at the start was
1383        studied, the wrong optimizing data was generated, leading to matching
1384        errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
1385        matching string must start with a or b or c. The correct conclusion for
1386        this pattern is that a match can start with any character.
1387    
1388    
1389    Version 4.4 13-Aug-03
1390    ---------------------
1391    
1392     1. In UTF-8 mode, a character class containing characters with values between
1393        127 and 255 was not handled correctly if the compiled pattern was studied.
1394        In fixing this, I have also improved the studying algorithm for such
1395        classes (slightly).
1396    
1397     2. Three internal functions had redundant arguments passed to them. Removal
1398        might give a very teeny performance improvement.
1399    
1400     3. Documentation bug: the value of the capture_top field in a callout is *one
1401        more than* the number of the hightest numbered captured substring.
1402    
1403     4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
1404        in incorrectly linking with a previously installed version. They now link
1405        explicitly with libpcre.la.
1406    
1407     5. configure.in no longer needs to recognize Cygwin specially.
1408    
1409     6. A problem in pcre.in for Windows platforms is fixed.
1410    
1411     7. If a pattern was successfully studied, and the -d (or /D) flag was given to
1412        pcretest, it used to include the size of the study block as part of its
1413        output. Unfortunately, the structure contains a field that has a different
1414        size on different hardware architectures. This meant that the tests that
1415        showed this size failed. As the block is currently always of a fixed size,
1416        this information isn't actually particularly useful in pcretest output, so
1417        I have just removed it.
1418    
1419     8. Three pre-processor statements accidentally did not start in column 1.
1420        Sadly, there are *still* compilers around that complain, even though
1421        standard C has not required this for well over a decade. Sigh.
1422    
1423     9. In pcretest, the code for checking callouts passed small integers in the
1424        callout_data field, which is a void * field. However, some picky compilers
1425        complained about the casts involved for this on 64-bit systems. Now
1426        pcretest passes the address of the small integer instead, which should get
1427        rid of the warnings.
1428    
1429    10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
1430        both compile and run time, and gives an error if an invalid UTF-8 sequence
1431        is found. There is a option for disabling this check in cases where the
1432        string is known to be correct and/or the maximum performance is wanted.
1433    
1434    11. In response to a bug report, I changed one line in Makefile.in from
1435    
1436            -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
1437        to
1438            -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
1439    
1440        to look similar to other lines, but I have no way of telling whether this
1441        is the right thing to do, as I do not use Windows. No doubt I'll get told
1442        if it's wrong...
1443    
1444    
1445    Version 4.3 21-May-03
1446    ---------------------
1447    
1448    1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
1449       Makefile.
1450    
1451    2. Some refactoring to improve the quality of the code:
1452    
1453       (i)   The utf8_table... variables are now declared "const".
1454    
1455       (ii)  The code for \cx, which used the "case flipping" table to upper case
1456             lower case letters, now just substracts 32. This is ASCII-specific,
1457             but the whole concept of \cx is ASCII-specific, so it seems
1458             reasonable.
1459    
1460       (iii) PCRE was using its character types table to recognize decimal and
1461             hexadecimal digits in the pattern. This is silly, because it handles
1462             only 0-9, a-f, and A-F, but the character types table is locale-
1463             specific, which means strange things might happen. A private
1464             table is now used for this - though it costs 256 bytes, a table is
1465             much faster than multiple explicit tests. Of course, the standard
1466             character types table is still used for matching digits in subject
1467             strings against \d.
1468    
1469       (iv)  Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
1470             ending in _t are). So I've renamed it as ESC_tee.
1471    
1472    3. The first argument for regexec() in the POSIX wrapper should have been
1473       defined as "const".
1474    
1475    4. Changed pcretest to use malloc() for its buffers so that they can be
1476       Electric Fenced for debugging.
1477    
1478    5. There were several places in the code where, in UTF-8 mode, PCRE would try
1479       to read one or more bytes before the start of the subject string. Often this
1480       had no effect on PCRE's behaviour, but in some circumstances it could
1481       provoke a segmentation fault.
1482    
1483    6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
1484       to try to read one or more bytes before the start of the subject string.
1485    
1486    7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
1487       UTF-8 support could misbehave in various ways if the subject string
1488       contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
1489       area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
1490       back over UTF-8 characters.)
1491    
1492    
1493    Version 4.2 14-Apr-03
1494    ---------------------
1495    
1496    1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
1497    
1498    2. Changes to the building process, supplied by Ronald Landheer-Cieslak
1499         [ON_WINDOWS]: new variable, "#" on non-Windows platforms
1500         [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
1501         [WIN_PREFIX]: new variable, "cyg" for Cygwin
1502         * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
1503           and BUILD_EXEEXT
1504         Note: automatic setting of the BUILD variables is not yet working
1505         set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
1506           compile-time but not at link-time
1507         [LINK]: use for linking executables only
1508         make different versions for Windows and non-Windows
1509         [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
1510           libraries
1511         [LINK_FOR_BUILD]: new variable
1512         [OBJEXT]: use throughout
1513         [EXEEXT]: use throughout
1514         <winshared>: new target
1515         <wininstall>: new target
1516         <dftables.o>: use native compiler
1517         <dftables>: use native linker
1518         <install>: handle Windows platform correctly
1519         <clean>: ditto
1520         <check>: ditto
1521         copy DLL to top builddir before testing
1522    
1523       As part of these changes, -no-undefined was removed again. This was reported
1524       to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
1525       in any case.
1526    
1527    3. Some tidies to get rid of compiler warnings:
1528    
1529       . In the match_data structure, match_limit was an unsigned long int, whereas
1530         match_call_count was an int. I've made them both unsigned long ints.
1531    
1532       . In pcretest the fact that a const uschar * doesn't automatically cast to
1533         a void * provoked a warning.
1534    
1535       . Turning on some more compiler warnings threw up some "shadow" variables
1536         and a few more missing casts.
1537    
1538    4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
1539       option, a class that contained a single character with a value between 128
1540       and 255 (e.g. /[\xFF]/) caused PCRE to crash.
1541    
1542    5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
1543       option, a class that contained several characters, but with at least one
1544       whose value was between 128 and 255 caused PCRE to crash.
1545    
1546    
1547    Version 4.1 12-Mar-03
1548    ---------------------
1549    
1550    1. Compiling with gcc -pedantic found a couple of places where casts were
1551    needed, and a string in dftables.c that was longer than standard compilers are
1552    required to support.
1553    
1554    2. Compiling with Sun's compiler found a few more places where the code could
1555    be tidied up in order to avoid warnings.
1556    
1557    3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
1558    first of these names is deprecated in the latest Autoconf in favour of the name
1559    CC_FOR_BUILD, because "host" is typically used to mean the system on which the
1560    compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
1561    analogy I have changed it to CFLAGS_FOR_BUILD.
1562    
1563    4. Added -no-undefined to the linking command in the Makefile, because this is
1564    apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
1565    linking step for the pcreposix library.
1566    
1567    5. PCRE was failing to diagnose the case of two named groups with the same
1568    name.
1569    
1570    6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
1571    literal character that is needed in the subject for a match, and scans along to
1572    ensure that it is present before embarking on the full matching process. This
1573    saves time in cases of nested unlimited repeats that are never going to match.
1574    Problem: the scan can take a lot of time if the subject is very long (e.g.
1575    megabytes), thus penalizing straightforward matches. It is now done only if the
1576    amount of subject to be scanned is less than 1000 bytes.
1577    
1578    7. A lesser problem with the same optimization is that it was recording the
1579    first character of an anchored pattern as "needed", thus provoking a search
1580    right along the subject, even when the first match of the pattern was going to
1581    fail. The "needed" character is now not set for anchored patterns, unless it
1582    follows something in the pattern that is of non-fixed length. Thus, it still
1583    fulfils its original purpose of finding quick non-matches in cases of nested
1584    unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
1585    
1586    
1587    Version 4.0 17-Feb-03
1588    ---------------------
1589    
1590    1. If a comment in an extended regex that started immediately after a meta-item
1591    extended to the end of string, PCRE compiled incorrect data. This could lead to
1592    all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
1593    
1594    2. Moved to autoconf 2.53 and libtool 1.4.2.
1595    
1596    3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
1597    the special perltest8 script is no longer needed - all the tests can be run
1598    from a single perltest script.
1599    
1600    4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
1601    by \s. It has now been removed in PCRE. This means it isn't recognized as
1602    whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
1603    class [:space:] *does* include VT, thereby creating a mess.
1604    
1605    5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
1606    space and tab.
1607    
1608    6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
1609    its new features into the main test script, reducing the number of scripts.
1610    
1611    7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
1612    were backward compatible, and made the (?i) apply to the whole pattern, as if
1613    /i were given. Now it behaves more logically, and applies the option setting
1614    only to what follows. PCRE has been changed to follow suit. However, if it
1615    finds options settings right at the start of the pattern, it extracts them into
1616    the global options, as before. Thus, they show up in the info data.
1617    
1618    8. Added support for the \Q...\E escape sequence. Characters in between are
1619    treated as literals. This is slightly different from Perl in that $ and @ are
1620    also handled as literals inside the quotes. In Perl, they will cause variable
1621    interpolation. Note the following examples:
1622    
1623        Pattern            PCRE matches      Perl matches
1624    
1625        \Qabc$xyz\E        abc$xyz           abc followed by the contents of $xyz
1626        \Qabc\$xyz\E       abc\$xyz          abc\$xyz
1627        \Qabc\E\$\Qxyz\E   abc$xyz           abc$xyz
1628    
1629    For compatibility with Perl, \Q...\E sequences are recognized inside character
1630    classes as well as outside them.
1631    
1632    9. Re-organized 3 code statements in pcretest to avoid "overflow in
1633    floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
1634    (size_t) cast to one statement in pcretest and one in pcreposix to avoid
1635    signed/unsigned warnings.
1636    
1637    10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
1638    option for pcretest, so I've replaced it by a simple function that does just
1639    that job.
1640    
1641    11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
1642    "pcregrep -".
1643    
1644    12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
1645    Java package. This provides some syntactic sugar for simple cases of what my
1646    documentation calls "once-only subpatterns". A pattern such as x*+ is the same
1647    as (?>x*). In other words, if what is inside (?>...) is just a single repeated
1648    item, you can use this simplified notation. Note that only makes sense with
1649    greedy quantifiers. Consequently, the use of the possessive quantifier forces
1650    greediness, whatever the setting of the PCRE_UNGREEDY option.
1651    
1652    13. A change of greediness default within a pattern was not taking effect at
1653    the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
1654    subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
1655    was abstracted outside.
1656    
1657    14. PCRE now supports the \G assertion. It is true when the current matching
1658    position is at the start point of the match. This differs from \A when the
1659    starting offset is non-zero. Used with the /g option of pcretest (or similar
1660    code), it works in the same way as it does for Perl's /g option. If all
1661    alternatives of a regex begin with \G, the expression is anchored to the start
1662    match position, and the "anchored" flag is set in the compiled expression.
1663    
1664    15. Some bugs concerning the handling of certain option changes within patterns
1665    have been fixed. These applied to options other than (?ims). For example,
1666    "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
1667    been the other way round. Some of this was related to change 7 above.
1668    
1669    16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
1670    features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
1671    and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
1672    POSIX classes only within a class (e.g. /[[:alpha:]]/).
1673    
1674    17. Added support for Perl's \C escape. This matches one byte, even in UTF8
1675    mode. Unlike ".", it always matches newline, whatever the setting of
1676    PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
1677    assertions. Perl allows it, but it doesn't (in general) work because it can't
1678    calculate the length of the lookbehind. At least, that's the case for Perl
1679    5.8.0 - I've been told they are going to document that it doesn't work in
1680    future.
1681    
1682    18. Added an error diagnosis for escapes that PCRE does not support: these are
1683    \L, \l, \N, \P, \p, \U, \u, and \X.
1684    
1685    19. Although correctly diagnosing a missing ']' in a character class, PCRE was
1686    reading past the end of the pattern in cases such as /[abcd/.
1687    
1688    20. PCRE was getting more memory than necessary for patterns with classes that
1689    contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
1690    
1691    21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
1692    compiling PCRE for use with Virtual Pascal.
1693    
1694    22. Small fix to the Makefile to make it work properly if the build is done
1695    outside the source tree.
1696    
1697    23. Added a new extension: a condition to go with recursion. If a conditional
1698    subpattern starts with (?(R) the "true" branch is used if recursion has
1699    happened, whereas the "false" branch is used only at the top level.
1700    
1701    24. When there was a very long string of literal characters (over 255 bytes
1702    without UTF support, over 250 bytes with UTF support), the computation of how
1703    much memory was required could be incorrect, leading to segfaults or other
1704    strange effects.
1705    
1706    25. PCRE was incorrectly assuming anchoring (either to start of subject or to
1707    start of line for a non-DOTALL pattern) when a pattern started with (.*) and
1708    there was a subsequent back reference to those brackets. This meant that, for
1709    example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
1710    possible to check for precisely this case. All we can do is abandon the
1711    optimization if .* occurs inside capturing brackets when there are any back
1712    references whatsoever. (See below for a better fix that came later.)
1713    
1714    26. The handling of the optimization for finding the first character of a
1715    non-anchored pattern, and for finding a character that is required later in the
1716    match were failing in some cases. This didn't break the matching; it just
1717    failed to optimize when it could. The way this is done has been re-implemented.
1718    
1719    27. Fixed typo in error message for invalid (?R item (it said "(?p").
1720    
1721    28. Added a new feature that provides some of the functionality that Perl
1722    provides with (?{...}). The facility is termed a "callout". The way it is done
1723    in PCRE is for the caller to provide an optional function, by setting
1724    pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
1725    global variable. By default it is unset, which disables all calling out. To get
1726    the function called, the regex must include (?C) at appropriate points. This
1727    is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
1728    This provides a means of identifying different callout points. When PCRE
1729    reaches such a point in the regex, if pcre_callout has been set, the external
1730    function is called. It is provided with data in a structure called
1731    pcre_callout_block, which is defined in pcre.h. If the function returns 0,
1732    matching continues; if it returns a non-zero value, the match at the current
1733    point fails. However, backtracking will occur if possible. [This was changed
1734    later and other features added - see item 49 below.]
1735    
1736    29. pcretest is upgraded to test the callout functionality. It provides a
1737    callout function that displays information. By default, it shows the start of
1738    the match and the current position in the text. There are some new data escapes
1739    to vary what happens:
1740    
1741        \C+         in addition, show current contents of captured substrings
1742        \C-         do not supply a callout function
1743        \C!n        return 1 when callout number n is reached
1744        \C!n!m      return 1 when callout number n is reached for the mth time
1745    
1746    30. If pcregrep was called with the -l option and just a single file name, it
1747    output "<stdin>" if a match was found, instead of the file name.
1748    
1749    31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
1750    slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
1751    pcre_exec(). This saves a malloc/free per call. The default value of
1752    POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
1753    when configuring.
1754    
1755    32. The default maximum size of a compiled pattern is 64K. There have been a
1756    few cases of people hitting this limit. The code now uses macros to handle the
1757    storing of links as offsets within the compiled pattern. It defaults to 2-byte
1758    links, but this can be changed to 3 or 4 bytes by --with-link-size when
1759    configuring. Tests 2 and 5 work only with 2-byte links because they output
1760    debugging information about compiled patterns.
1761    
1762    33. Internal code re-arrangements:
1763    
1764    (a) Moved the debugging function for printing out a compiled regex into
1765        its own source file (printint.c) and used #include to pull it into
1766        pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
1767        separate copies.
1768    
1769    (b) Defined the list of op-code names for debugging as a macro in
1770        internal.h so that it is next to the definition of the opcodes.
1771    
1772    (c) Defined a table of op-code lengths for simpler skipping along compiled
1773        code. This is again a macro in internal.h so that it is next to the
1774        definition of the opcodes.
1775    
1776    34. Added support for recursive calls to individual subpatterns, along the
1777    lines of Robin Houston's patch (but implemented somewhat differently).
1778    
1779    35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
1780    allow it to read and process whole directories in Win32. This code was
1781    contributed by Lionel Fourquaux; it has not been tested by me.
1782    
1783    36. Added support for named subpatterns. The Python syntax (?P<name>...) is
1784    used to name a group. Names consist of alphanumerics and underscores, and must
1785    be unique. Back references use the syntax (?P=name) and recursive calls use
1786    (?P>name) which is a PCRE extension to the Python extension. Groups still have
1787    numbers. The function pcre_fullinfo() can be used after compilation to extract
1788    a name/number map. There are three relevant calls:
1789    
1790      PCRE_INFO_NAMEENTRYSIZE        yields the size of each entry in the map
1791      PCRE_INFO_NAMECOUNT            yields the number of entries
1792      PCRE_INFO_NAMETABLE            yields a pointer to the map.
1793    
1794    The map is a vector of fixed-size entries. The size of each entry depends on
1795    the length of the longest name used. The first two bytes of each entry are the
1796    group number, most significant byte first. There follows the corresponding
1797    name, zero terminated. The names are in alphabetical order.
1798    
1799    37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
1800    case instead of 255. Making it the same both with and without UTF-8 support
1801    means that the same test output works with both.
1802    
1803    38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
1804    calling malloc() with a zero argument.
1805    
1806    39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
1807    optimization. I've improved things by keeping a bitmap of backreferences with
1808    numbers 1-31 so that if .* occurs inside capturing brackets that are not in
1809    fact referenced, the optimization can be applied. It is unlikely that a
1810    relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
1811    the match to follow \n) will appear inside brackets with a number greater than
1812    31, but if it does, any back reference > 31 suppresses the optimization.
1813    
1814    40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
1815    of disabling numbered capturing parentheses. Any opening parenthesis that is
1816    not followed by ? behaves as if it were followed by ?: but named parentheses
1817    can still be used for capturing (and they will acquire numbers in the usual
1818    way).
1819    
1820    41. Redesigned the return codes from the match() function into yes/no/error so
1821    that errors can be passed back from deep inside the nested calls. A malloc
1822    failure while inside a recursive subpattern call now causes the
1823    PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
1824    
1825    42. It is now possible to set a limit on the number of times the match()
1826    function is called in a call to pcre_exec(). This facility makes it possible to
1827    limit the amount of recursion and backtracking, though not in a directly
1828    obvious way, because the match() function is used in a number of different
1829    circumstances. The count starts from zero for each position in the subject
1830    string (for non-anchored patterns). The default limit is, for compatibility, a
1831    large number, namely 10 000 000. You can change this in two ways:
1832    
1833    (a) When configuring PCRE before making, you can use --with-match-limit=n
1834        to set a default value for the compiled library.
1835    
1836    (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
1837        a different value is set. See 45 below.
1838    
1839    If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
1840    
1841    43. Added a new function pcre_config(int, void *) to enable run-time extraction
1842    of things that can be changed at compile time. The first argument specifies
1843    what is wanted and the second points to where the information is to be placed.
1844    The current list of available information is:
1845    
1846      PCRE_CONFIG_UTF8
1847    
1848    The output is an integer that is set to one if UTF-8 support is available;
1849    otherwise it is set to zero.
1850    
1851      PCRE_CONFIG_NEWLINE
1852    
1853    The output is an integer that it set to the value of the code that is used for
1854    newline. It is either LF (10) or CR (13).
1855    
1856      PCRE_CONFIG_LINK_SIZE
1857    
1858    The output is an integer that contains the number of bytes used for internal
1859    linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
1860    
1861      PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
1862    
1863    The output is an integer that contains the threshold above which the POSIX
1864    interface uses malloc() for output vectors. See item 31 above.
1865    
1866      PCRE_CONFIG_MATCH_LIMIT
1867    
1868    The output is an unsigned integer that contains the default limit of the number
1869    of match() calls in a pcre_exec() execution. See 42 above.
1870    
1871    44. pcretest has been upgraded by the addition of the -C option. This causes it
1872    to extract all the available output from the new pcre_config() function, and to
1873    output it. The program then exits immediately.
1874    
1875    45. A need has arisen to pass over additional data with calls to pcre_exec() in
1876    order to support additional features. One way would have been to define
1877    pcre_exec2() (for example) with extra arguments, but this would not have been
1878    extensible, and would also have required all calls to the original function to
1879    be mapped to the new one. Instead, I have chosen to extend the mechanism that
1880    is used for passing in "extra" data from pcre_study().
1881    
1882    The pcre_extra structure is now exposed and defined in pcre.h. It currently
1883    contains the following fields:
1884    
1885      flags         a bitmap indicating which of the following fields are set
1886      study_data    opaque data from pcre_study()
1887      match_limit   a way of specifying a limit on match() calls for a specific
1888                      call to pcre_exec()
1889      callout_data  data for callouts (see 49 below)
1890    
1891    The flag bits are also defined in pcre.h, and are
1892    
1893      PCRE_EXTRA_STUDY_DATA
1894      PCRE_EXTRA_MATCH_LIMIT
1895      PCRE_EXTRA_CALLOUT_DATA
1896    
1897    The pcre_study() function now returns one of these new pcre_extra blocks, with
1898    the actual study data pointed to by the study_data field, and the
1899    PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
1900    before. That is, this change is entirely upwards-compatible and requires no
1901    change to existing code.
1902    
1903    If you want to pass in additional data to pcre_exec(), you can either place it
1904    in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
1905    block.
1906    
1907    46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
1908    data string contains the escape sequence \M, pcretest calls pcre_exec() several
1909    times with different match limits, until it finds the minimum value needed for
1910    pcre_exec() to complete. The value is then output. This can be instructive; for
1911    most simple matches the number is quite small, but for pathological cases it
1912    gets very large very quickly.
1913    
1914    47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
1915    returns the size of the data block pointed to by the study_data field in a
1916    pcre_extra block, that is, the value that was passed as the argument to
1917    pcre_malloc() when PCRE was getting memory in which to place the information
1918    created by pcre_study(). The fourth argument should point to a size_t variable.
1919    pcretest has been extended so that this information is shown after a successful
1920    pcre_study() call when information about the compiled regex is being displayed.
1921    
1922    48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
1923    because what follows is always an absolute path. (Later: it turns out that this
1924    is more than cosmetic for MinGW, because it doesn't like empty path
1925    components.)
1926    
1927    49. Some changes have been made to the callout feature (see 28 above):
1928    
1929    (i)  A callout function now has three choices for what it returns:
1930    
1931           0  =>  success, carry on matching
1932         > 0  =>  failure at this point, but backtrack if possible
1933         < 0  =>  serious error, return this value from pcre_exec()
1934    
1935         Negative values should normally be chosen from the set of PCRE_ERROR_xxx
1936         values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
1937         "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
1938         use by callout functions. It will never be used by PCRE itself.
1939    
1940    (ii) The pcre_extra structure (see 45 above) has a void * field called
1941         callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
1942         pcre_callout_block structure has a field of the same name. The contents of
1943         the field passed in the pcre_extra structure are passed to the callout
1944         function in the corresponding field in the callout block. This makes it
1945         easier to use the same callout-containing regex from multiple threads. For
1946         testing, the pcretest program has a new data escape
1947    
1948           \C*n        pass the number n (may be negative) as callout_data
1949    
1950         If the callout function in pcretest receives a non-zero value as
1951         callout_data, it returns that value.
1952    
1953    50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
1954    there were some redundant $(CFLAGS) in commands that are now specified as
1955    $(LINK), which already includes $(CFLAGS).
1956    
1957    51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
1958    has been compiled with UTF-8 support *and* pcre_compile() has been compiled
1959    with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
1960    one-byte characters throughout. Note that case-insensitive matching applies
1961    only to characters whose values are less than 256. PCRE doesn't support the
1962    notion of cases for higher-valued characters.
1963    
1964    (i)   A character class whose characters are all within 0-255 is handled as
1965          a bit map, and the map is inverted for negative classes. Previously, a
1966          character > 255 always failed to match such a class; however it should
1967          match if the class was a negative one (e.g. [^ab]). This has been fixed.
1968    
1969    (ii)  A negated character class with a single character < 255 is coded as
1970          "not this character" (OP_NOT). This wasn't working properly when the test
1971          character was multibyte, either singly or repeated.
1972    
1973    (iii) Repeats of multibyte characters are now handled correctly in UTF-8
1974          mode, for example: \x{100}{2,3}.
1975    
1976    (iv)  The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
1977          singly or repeated) now correctly test multibyte characters. However,
1978          PCRE doesn't recognize any characters with values greater than 255 as
1979          digits, spaces, or word characters. Such characters always match \D, \S,
1980          and \W, and never match \d, \s, or \w.
1981    
1982    (v)   Classes may now contain characters and character ranges with values
1983          greater than 255. For example: [ab\x{100}-\x{400}].
1984    
1985    (vi)  pcregrep now has a --utf-8 option (synonym -u) which makes it call
1986          PCRE in UTF-8 mode.
1987    
1988    52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
1989    PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
1990    retained for backwards compatibility. (Note that LASTLITERAL is also a byte
1991    value.)
1992    
1993    53. The single man page has become too large. I have therefore split it up into
1994    a number of separate man pages. These also give rise to individual HTML pages;
1995    these are now put in a separate directory, and there is an index.html page that
1996    lists them all. Some hyperlinking between the pages has been installed.
1997    
1998    54. Added convenience functions for handling named capturing parentheses.
1999    
2000    55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
2001    aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
2002    true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
2003    are faulted.
2004    
2005    56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
2006    calling configure. These values are used when compiling the dftables.c program
2007    which is run to generate the source of the default character tables. They
2008    default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
2009    you will need to set these values.
2010    
2011    57. Updated the building process for Windows DLL, as provided by Fred Cox.
2012    
2013    
2014    Version 3.9 02-Jan-02
2015    ---------------------
2016    
2017    1. A bit of extraneous text had somehow crept into the pcregrep documentation.
2018    
2019    2. If --disable-static was given, the building process failed when trying to
2020    build pcretest and pcregrep. (For some reason it was using libtool to compile
2021    them, which is not right, as they aren't part of the library.)
2022    
2023    
2024    Version 3.8 18-Dec-01
2025    ---------------------
2026    
2027    1. The experimental UTF-8 code was completely screwed up. It was packing the
2028    bytes in the wrong order. How dumb can you get?
2029    
2030    
2031    Version 3.7 29-Oct-01
2032    ---------------------
2033    
2034    1. In updating pcretest to check change 1 of version 3.6, I screwed up.
2035    This caused pcretest, when used on the test data, to segfault. Unfortunately,
2036    this didn't happen under Solaris 8, where I normally test things.
2037    
2038    2. The Makefile had to be changed to make it work on BSD systems, where 'make'
2039    doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
2040    isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
2041    this fix an hour or so after the initial 3.7 release.)
2042    
2043    
2044    Version 3.6 23-Oct-01
2045    ---------------------
2046    
2047    1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
2048    offsets passed as NULL with zero offset count.
2049    
2050    2. The config.guess and config.sub files had not been updated when I moved to
2051    the latest autoconf.
2052    
2053    
2054    Version 3.5 15-Aug-01
2055    ---------------------
2056    
2057    1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
2058    had been forgotten.
2059    
2060    2. By using declared but undefined structures, we can avoid using "void"
2061    definitions in pcre.h while keeping the internal definitions of the structures
2062    private.
2063    
2064    3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
2065    user point of view, this means that both static and shared libraries are built
2066    by default, but this can be individually controlled. More of the work of
2067    handling this static/shared cases is now inside libtool instead of PCRE's make
2068    file.
2069    
2070    4. The pcretest utility is now installed along with pcregrep because it is
2071    useful for users (to test regexs) and by doing this, it automatically gets
2072    relinked by libtool. The documentation has been turned into a man page, so
2073    there are now .1, .txt, and .html versions in /doc.
2074    
2075    5. Upgrades to pcregrep:
2076       (i)   Added long-form option names like gnu grep.
2077       (ii)  Added --help to list all options with an explanatory phrase.
2078       (iii) Added -r, --recursive to recurse into sub-directories.
2079       (iv)  Added -f, --file to read patterns from a file.
2080    
2081    6. pcre_exec() was referring to its "code" argument before testing that
2082    argument for NULL (and giving an error if it was NULL).
2083    
2084    7. Upgraded Makefile.in to allow for compiling in a different directory from
2085    the source directory.
2086    
2087    8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
2088    options bits, the pointer it was passed was to an int instead of to an unsigned
2089    long int. This mattered only on 64-bit systems.
2090    
2091    9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
2092    generated) instead of pcre.in, which it its source. Also made the same change
2093    in several of the .c files.
2094    
2095    10. A new release of gcc defines printf() as a macro, which broke pcretest
2096    because it had an ifdef in the middle of a string argument for printf(). Fixed
2097    by using separate calls to printf().
2098    
2099    11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
2100    script, to force use of CR or LF instead of \n in the source. On non-Unix
2101    systems, the value can be set in config.h.
2102    
2103    12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
2104    absolute limit. Changed the text of the error message to make this clear, and
2105    likewise updated the man page.
2106    
2107    13. The limit of 99 on the number of capturing subpatterns has been removed.
2108    The new limit is 65535, which I hope will not be a "real" limit.
2109    
2110    
2111    Version 3.4 22-Aug-00
2112    ---------------------
2113    
2114    1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *.
2115    
2116    2. Diagnose condition (?(0) as an error instead of crashing on matching.
2117    
2118    
2119    Version 3.3 01-Aug-00
2120    ---------------------
2121    
2122    1. If an octal character was given, but the value was greater than \377, it
2123    was not getting masked to the least significant bits, as documented. This could
2124    lead to crashes in some systems.
2125    
2126    2. Perl 5.6 (if not earlier versions) accepts classes like [a-\d] and treats
2127    the hyphen as a literal. PCRE used to give an error; it now behaves like Perl.
2128    
2129    3. Added the functions pcre_free_substring() and pcre_free_substring_list().
2130    These just pass their arguments on to (pcre_free)(), but they are provided
2131    because some uses of PCRE bind it to non-C systems that can call its functions,
2132    but cannot call free() or pcre_free() directly.
2133    
2134    4. Add "make test" as a synonym for "make check". Corrected some comments in
2135    the Makefile.
2136    
2137    5. Add $(DESTDIR)/ in front of all the paths in the "install" target in the
2138    Makefile.
2139    
2140    6. Changed the name of pgrep to pcregrep, because Solaris has introduced a
2141    command called pgrep for grepping around the active processes.
2142    
2143    7. Added the beginnings of support for UTF-8 character strings.
2144    
2145    8. Arranged for the Makefile to pass over the settings of CC, CFLAGS, and
2146    RANLIB to ./ltconfig so that they are used by libtool. I think these are all
2147    the relevant ones. (AR is not passed because ./ltconfig does its own figuring
2148    out for the ar command.)
2149    
2150    
2151    Version 3.2 12-May-00
2152    ---------------------
2153    
2154    This is purely a bug fixing release.
2155    
2156    1. If the pattern /((Z)+|A)*/ was matched agained ZABCDEFG it matched Z instead
2157    of ZA. This was just one example of several cases that could provoke this bug,
2158    which was introduced by change 9 of version 2.00. The code for breaking
2159    infinite loops after an iteration that matches an empty string was't working
2160    correctly.
2161    
2162    2. The pcretest program was not imitating Perl correctly for the pattern /a*/g
2163    when matched against abbab (for example). After matching an empty string, it
2164    wasn't forcing anchoring when setting PCRE_NOTEMPTY for the next attempt; this
2165    caused it to match further down the string than it should.
2166    
2167    3. The code contained an inclusion of sys/types.h. It isn't clear why this
2168    was there because it doesn't seem to be needed, and it causes trouble on some
2169    systems, as it is not a Standard C header. It has been removed.
2170    
2171    4. Made 4 silly changes to the source to avoid stupid compiler warnings that
2172    were reported on the Macintosh. The changes were from
2173    
2174      while ((c = *(++ptr)) != 0 && c != '\n');
2175    to
2176      while ((c = *(++ptr)) != 0 && c != '\n') ;
2177    
2178    Totally extraordinary, but if that's what it takes...
2179    
2180    5. PCRE is being used in one environment where neither memmove() nor bcopy() is
2181    available. Added HAVE_BCOPY and an autoconf test for it; if neither
2182    HAVE_MEMMOVE nor HAVE_BCOPY is set, use a built-in emulation function which
2183    assumes the way PCRE uses memmove() (always moving upwards).
2184    
2185    6. PCRE is being used in one environment where strchr() is not available. There
2186    was only one use in pcre.c, and writing it out to avoid strchr() probably gives
2187    faster code anyway.
2188    
2189    
2190    Version 3.1 09-Feb-00
2191    ---------------------
2192    
2193    The only change in this release is the fixing of some bugs in Makefile.in for
2194    the "install" target:
2195    
2196    (1) It was failing to install pcreposix.h.
2197    
2198    (2) It was overwriting the pcre.3 man page with the pcreposix.3 man page.
2199    
2200    
2201    Version 3.0 01-Feb-00
2202    ---------------------
2203    
2204    1. Add support for the /+ modifier to perltest (to output $` like it does in
2205    pcretest).
2206    
2207    2. Add support for the /g modifier to perltest.
2208    
2209    3. Fix pcretest so that it behaves even more like Perl for /g when the pattern
2210    matches null strings.
2211    
2212    4. Fix perltest so that it doesn't do unwanted things when fed an empty
2213    pattern. Perl treats empty patterns specially - it reuses the most recent
2214    pattern, which is not what we want. Replace // by /(?#)/ in order to avoid this
2215    effect.
2216    
2217    5. The POSIX interface was broken in that it was just handing over the POSIX
2218    captured string vector to pcre_exec(), but (since release 2.00) PCRE has
2219    required a bigger vector, with some working space on the end. This means that
2220    the POSIX wrapper now has to get and free some memory, and copy the results.
2221    
2222    6. Added some simple autoconf support, placing the test data and the
2223    documentation in separate directories, re-organizing some of the
2224    information files, and making it build pcre-config (a GNU standard). Also added
2225    libtool support for building PCRE as a shared library, which is now the
2226    default.
2227    
2228    7. Got rid of the leading zero in the definition of PCRE_MINOR because 08 and
2229    09 are not valid octal constants. Single digits will be used for minor values
2230    less than 10.
2231    
2232    8. Defined REG_EXTENDED and REG_NOSUB as zero in the POSIX header, so that
2233    existing programs that set these in the POSIX interface can use PCRE without
2234    modification.
2235    
2236    9. Added a new function, pcre_fullinfo() with an extensible interface. It can
2237    return all that pcre_info() returns, plus additional data. The pcre_info()
2238    function is retained for compatibility, but is considered to be obsolete.
2239    
2240    10. Added experimental recursion feature (?R) to handle one common case that
2241    Perl 5.6 will be able to do with (?p{...}).
2242    
2243    11. Added support for POSIX character classes like [:alpha:], which Perl is
2244    adopting.
2245    
2246    
2247    Version 2.08 31-Aug-99
2248    ----------------------
2249    
2250    1. When startoffset was not zero and the pattern began with ".*", PCRE was not
2251    trying to match at the startoffset position, but instead was moving forward to
2252    the next newline as if a previous match had failed.
2253    
2254    2. pcretest was not making use of PCRE_NOTEMPTY when repeating for /g and /G,
2255    and could get into a loop if a null string was matched other than at the start
2256    of the subject.
2257    
2258    3. Added definitions of PCRE_MAJOR and PCRE_MINOR to pcre.h so the version can
2259    be distinguished at compile time, and for completeness also added PCRE_DATE.
2260    
2261    5. Added Paul Sokolovsky's minor changes to make it easy to compile a Win32 DLL
2262    in GnuWin32 environments.
2263    
2264    
2265    Version 2.07 29-Jul-99
2266    ----------------------
2267    
2268    1. The documentation is now supplied in plain text form and HTML as well as in
2269    the form of man page sources.
2270    
2271    2. C++ compilers don't like assigning (void *) values to other pointer types.
2272    In particular this affects malloc(). Although there is no problem in Standard
2273    C, I've put in casts to keep C++ compilers happy.
2274    
2275    3. Typo on pcretest.c; a cast of (unsigned char *) in the POSIX regexec() call
2276    should be (const char *).
2277    
2278    4. If NOPOSIX is defined, pcretest.c compiles without POSIX support. This may
2279    be useful for non-Unix systems who don't want to bother with the POSIX stuff.
2280    However, I haven't made this a standard facility. The documentation doesn't
2281    mention it, and the Makefile doesn't support it.
2282    
2283    5. The Makefile now contains an "install" target, with editable destinations at
2284    the top of the file. The pcretest program is not installed.
2285    
2286    6. pgrep -V now gives the PCRE version number and date.
2287    
2288    7. Fixed bug: a zero repetition after a literal string (e.g. /abcde{0}/) was
2289    causing the entire string to be ignored, instead of just the last character.
2290    
2291    8. If a pattern like /"([^\\"]+|\\.)*"/ is applied in the normal way to a
2292    non-matching string, it can take a very, very long time, even for strings of
2293    quite modest length, because of the nested recursion. PCRE now does better in
2294    some of these cases. It does this by remembering the last required literal
2295    character in the pattern, and pre-searching the subject to ensure it is present
2296    before running the real match. In other words, it applies a heuristic to detect
2297    some types of certain failure quickly, and in the above example, if presented
2298    with a string that has no trailing " it gives "no match" very quickly.
2299    
2300    9. A new runtime option PCRE_NOTEMPTY causes null string matches to be ignored;
2301    other alternatives are tried instead.
2302    
2303    
2304    Version 2.06 09-Jun-99
2305    ----------------------
2306    
2307    1. Change pcretest's output for amount of store used to show just the code
2308    space, because the remainder (the data block) varies in size between 32-bit and
2309    64-bit systems.
2310    
2311    2. Added an extra argument to pcre_exec() to supply an offset in the subject to
2312    start matching at. This allows lookbehinds to work when searching for multiple
2313    occurrences in a string.
2314    
2315    3. Added additional options to pcretest for testing multiple occurrences:
2316    
2317       /+   outputs the rest of the string that follows a match
2318       /g   loops for multiple occurrences, using the new startoffset argument
2319       /G   loops for multiple occurrences by passing an incremented pointer
2320    
2321    4. PCRE wasn't doing the "first character" optimization for patterns starting
2322    with \b or \B, though it was doing it for other lookbehind assertions. That is,
2323    it wasn't noticing that a match for a pattern such as /\bxyz/ has to start with
2324    the letter 'x'. On long subject strings, this gives a significant speed-up.
2325    
2326    
2327    Version 2.05 21-Apr-99
2328    ----------------------
2329    
2330    1. Changed the type of magic_number from int to long int so that it works
2331    properly on 16-bit systems.
2332    
2333    2. Fixed a bug which caused patterns starting with .* not to work correctly
2334    when the subject string contained newline characters. PCRE was assuming
2335    anchoring for such patterns in all cases, which is not correct because .* will
2336    not pass a newline unless PCRE_DOTALL is set. It now assumes anchoring only if
2337    DOTALL is set at top level; otherwise it knows that patterns starting with .*
2338    must be retried after every newline in the subject.
2339    
2340    
2341  Version 2.04 18-Feb-99  Version 2.04 18-Feb-99
2342  ----------------------  ----------------------

Legend:
Removed from v.31  
changed lines
  Added in v.175

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12